Hardware-in-Loop simulation of a White-Fi system

utilizing the TV White Spaces in India

Enabling cost-effective rural broadband to ensure total Internet adoption

Bharath Keshavamurthy, Abhay Narasimha, Asif Ahmad A S, Mahesh N, Suma M N

Department of Electronics and Communication Engineering,
BMS College of Engineering, Bull Temple Road, Bengaluru, India.
Email: {bharathkeshavamurthy.811,abhaynarasimha1,asif949,mahilion01}@[Link],[Link]@[Link]

Abstract—“The world-wide web still isn’t all that worldwide”.

This line by the Washington Post is a staggering fact- hard to I. INTRODUCTION
believe, but unfortunately very true. An exhaustive study by In the past few decades, the ‘phenomenon of the Internet’
McKinsey and Company [1] regarding the challenges of has proliferated from an embryonic technology to an extremely
worldwide Internet connectivity reveals a surprising reality: relevant tool that is the primary driving force behind thousands
Almost 4 billion people still aren’t connected to the Internet and
of innovative industries, applications and services. The impact
this fact amounts to a huge disparity between the technologically
privileged and their underprivileged counterparts. Around 1
of the Internet is mind-blowing. The Internet has significant
billion people in India are not connected to the Internet and this impacts on a country’s social, economic, cultural, educational
makes up for almost a quarter of the world’s offline population. and political progress. It has also generated massive socio-
The majority of people still unconnected to the Internet reside in cultural changes by connecting individuals and communities,
the country’s largely neglected rural communities and this puts providing seamless access to a treasure chest of seemingly
them at a considerable disadvantage- They are denied quality infinite content while promoting a certain level of transparency.
healthcare, education and government services which would However, the cold reality is that not all regions have exploited
undoubtedly make their lives much easier. Sheer neglect for our the enormous potential of the Internet to the same extent. In
villages, technical difficulties faced while deploying connectivity this article we look into some of the factors impeding global
services to seemingly geographically hard-to-reach localities and deployment of this iconic tool and suitable solutions to realize
monetary obstacles are the primary reasons for the lack of the much talked about dream of a Networked Society.
broadband penetration in the country. There is a tremendous
technological gap between the country’s urban and rural areas. Since 2004, the global online population increased to
While, people in the metropolitan areas enjoy seamless around 2.7 billion, propelled by numerous trends such as
connectivity, people in the villages and suburbs lack basic urbanization, increasingly plummeting device and network data
infrastructure to facilitate fundamental connectivity. In this era plan costs, a thriving middle class, increased utility of the
of Networked Society, the unconnected people suffer a great deal Internet, expansion of mobile network coverage and increased
in terms of a plethora of privileges their connected counterparts mobile Internet adoption. Considering the recent projections
enjoy. In order to bridge the connectivity divide in India we have casting light on an enormous increase in online activity, around
proposed an innovative long-range broadband service that aims 600 million people are expected to join in and become a part of
to reach the otherwise unreachable rural areas of the country. the Internet ecosystem by 2017. Although these numbers sound
The proposed solution involves the controlled use of the white excellent, they are far from it. Even with this trajectory, four
spaces in the terrestrial TV spectrum to provide Wi-Fi enabled and half billion people will still be deprived of the resources
connectivity, which would ensure cheap, efficient and relevant offered by the Internet. The socio-economic characteristics,
Internet services to the country’s rural offline population. “White demographic profiles and short-sighted mindsets of the current
Spaces” are segments/regions of the spectrum licensed by
offline population makes it highly unlikely that these
competent authorities, which primary users (licensees) do not use
individuals will become a part of a larger, virtually connected
all of the time or in all geographic locations. Several
communications entities and network operators are planning to
community, solely due to these trends which have fuelled
allow unlicensed use of these unused bands provided their adoption over the past decade.
primary use is not adversely affected. This enables the use of The exhaustive study by McKinsey & Company [1] has
these white spaces for applications which promote the nation’s revealed that around three-quarters of the global offline
development. population is concentrated in twenty countries and a majority
of these individuals are understandably from poor, rural,
Keywords—SCPI, EXA, EXG, IEEE 802.11b/g/n, VSA,
isolated and largely neglected communities. The obstacles
SystemVue, Spectrasys, IEEE 802.11af, UHF scanner, GDB,
RLSS, CAQ, RLQP, CSM, WSM, NFAP, WPC, NCO, DUC and
hindering nation-wide adoption of the Internet in India include,
DDC. Lack of suitable infrastructure (domain-specific and adjacent),
Lack of Digital literacy, Under-resourced education system,
Low income and affordability concerns, Disproportionately

978-1-5090-0893-3/16/$31.00 ©2016 IEEE

poor rural communities, High service costs, Unfavorable development in India and the tremendous technological gap
market structure, Lack of awareness, Lack of relevance and between the country’s urban and rural areas.
Lack of trusted logistics and payment systems.
With the advent of Wi-Fi (IEEE 802.11), there has been an
A country like India, according to the McKinsey & enormous boom in the usage of the Internet but however it is
Company [1], is categorized to possess ‘Medium to High mostly restricted to the urban and suburban areas of the
barriers’ to Internet adoption. India’s greatest challenge lies country. Many existing network connections suffer from last-
with the Incentives and Infrastructure barrier categories. mile connectivity issue. The wireless solutions based on Wi-Fi
Geographically isolated rural areas, Lack of nation-wide provide a cost-effective answer to the last-mile connectivity
mobile internet coverage, Limited spectrum availability and issue. However, Wi-Fi (IEEE 802.11b/g/n) operates in the
Ineffective national ICT strategies are the primary reasons for already congested ISM (Industrial, Scientific and Medical)
poor Internet penetration in India. band of 2.4GHz, where signal propagation characteristics are
not as good as in the sub-GHz bands such as the terrestrial TV
On the International front, numerous governments have bands. Most of the terrestrial TV bands are unused and
recognized the limitless potential of the Internet and the “Unused spectrum is a crime”.
benefits the citizens could reap through the scientific use of this
vast ocean of content and services. In order to cater to the We propose to use these unused TV bands in the terrestrial
offline population, several governments, non-governmental TV spectrum to provide Wi-Fi based broadband access which
agencies and IT firms have dived into the logistics of world- is cheaper and more effective than laying cable or optic fiber.
wide Internet adoption. Some of the services provided by these
concerned agencies/firms include increased public Wi-Fi Recent studies performed by CARE at IIT-Delhi [3]
access, increased mobile network coverage, awareness drives, suggest that nearly 85% of the terrestrial TV bands in the 470-
infrastructure development and reduced device and data plan 698MHz range go unused in urban Delhi while almost 95% of
costs. However, the deployment of the services promised by the TV bands are not utilized in the rural areas surrounding the
these agencies is painfully slow and would never keep up with NCR (National Capital Region). These statistics clearly suggest
the rate at which the offline and online population is growing. a tremendous scope for a system that facilitates the use of these
Hence, we need a comprehensive yet cost-effective approach to TV White Spaces to provide effective, yet affordable rural
tackle this global problem of insufficient Internet penetration Internet connectivity. We aim to ensure the development of
and we intend to do just that, with our innovative TVWS-based efficient, low-cost wireless solutions which exploit these
wireless communication solutions. abundantly available white spaces to provide cheap, affordable
broadband access to India’s villages. Using the under-utilized
Our proposal of fixed rural TVWS (TV White Spaces) terrestrial TV spectrum for broadband connectivity can be a
connectivity points providing seamless Wi-Fi based Internet paradigm-shifting innovation which enables a cost-effective
services over the TV White spaces aims to capitalize on the means of providing Internet access to lakhs of Indian villages.
government’s long-term goal of total Internet penetration. Our
system is a cheap, efficient and effective Internet access The IEEE 802.11af framework, referred to as ‘White-Fi’ or
scheme which is best suited for India’s rural communities ‘Super Wi-Fi’ is a wireless computer networking standard in
where TRAI’s extensive FTTx proposal would introduce the IEEE 802.11 family that allows WLAN operation in the TV
innumerable hassles and cost overruns. “Unused Spectrum is a White Space spectrum in the VHF and UHF bands in between
Crime”- We aim to reach the 800 million offline citizens and 54 and 790MHz (USA), 470-790MHz (Europe) and 470-
bring them into the Internet ecosystem, which would no doubt 698MHz (India). This standard was approved in 2014 and
facilitate a transformational impact on the country’s social, various cognitive radio technologies to facilitate the efficient
cultural, economic and political landscape. In this article, we use of these unused TV channels while limiting interference for
discuss the barriers impeding one hundred percent Internet primary users (ATV, DTV and digital microphones) are
adoption in India, our innovative solution to facilitate easy, currently being researched. In this project, our contribution to
efficient and cost-effective rural broadband deployment, a bridge the Digital Divide in India is three-fold.
survey to determine and establish the white spaces in the TV Firstly, we develop a novel spectrum scanning setup using a
spectrum and finally a Hardware-in-Loop simulation to spectrum analyzer and its related software (automation
validate our design compliance with IEEE 802.11af techniques and power analysis tools). Secondly, we present
specifications. analysis reports and surveys of the existing TV White Spaces
in the BMSCE, Bengaluru campus and develop models to
II. DESCRIPTION OF THE PROPOSED WHITE-FI PROTOTYPE output the most optimum TVWS bands for our use. Finally, we
develop a Hardware-in-loop processing platform complete with
A. Scope of the Proposed system the proposed UHF Scanner, 802.11b/g/n Wi-Fi source, UHF
Translator and the related DSP routines.
In many developing countries such as India, broadband
penetration in rural areas is staggeringly low due to the Although, there have been a few attempts to find viable
ridiculously high costs involved in laying cable and optic fiber solutions to capitalize on the potential applications of White
to every single household in the vast rural expanses of the Spaces in the TV spectrum, none of them succeed in
country. Many households in India’s villages do not have the developing a comprehensive framework suitable for the Indian
most basic Internet access such as browsing, OTT services, demographic and economic ecosystem. For instance,
web-search, etc. This brings out the pathetic state of References [4] and [8] present a quantitative evaluation of
TVWS availability in India. While [4] does present a making them highly favorable for long-range broadband
comprehensive and detailed analysis of TVWS availability in deployment, especially in the geographically isolated rural
India, it fails to provide techniques to tap into this enormous communities of India.
resource for the betterment of India’s rural population.
Reference [5] outlines observations and results from a series of One of the primary concerns of unlicensed TVWS
trials conducted in the United Kingdom with the commercial operations is to ensure non-interference with the incumbent
assistance of Ofcom. The results and discussions featured in [5] licensed users of the TV spectrum. The WSDs should learn
can serve as the foundation for similar research in India, from an approved Geo-location Data Base (GDB) regarding
facilitating efficient use of available White Spaces in the TV which channels are available and for what duration. Once
spectrum. Reference [6] presents a unique perspective on the operating in the GDB-approved channel, White Space Devices
design and development of Geo-Location Databases (in line are required not to interfere with the licensed users and
with FCC guidelines). The DynaWhite architecture proposed in moreover the White Space Devices should immediately stop
[6] facilitates the detection and distribution of dynamic TVWS transmitting in that channel, if the GDB tells them to stop.
information, based on data regarding both, the television The major differentiating factor between the IEEE 802.11af
transmitters and receivers. Another approach to enable standard and other 802.11 standards is the utilization of a Geo-
secondary wireless communications in the TV spectrum is Location Data Base (GDB) in the former. The GDB maintains
presented in [7] - allowing dynamic utilization of frequency geographically-characterized records of the permissible
resources between terrestrial TV broadcast and secondary frequencies and operational metrics of WSDs. The GDB
wireless broadband services. Reference [12] presents an Dependent (GDD) entities in the 802.11af network include the
analysis of downlink throughput performance of the IEEE GDD-Enabling station which is the equivalent of an Access
802.11af networking standard in a rural environment, taking Point and the GDD-Dependent station which is equivalent to a
into consideration real-world constraints such as availability STA (station) in the BSS (Basic Service Set) architecture. The
and co-channel interference. The promising results depicted in GDD-Dependent station obtains the permissible frequencies
[8] further strengthen our cause to ensure full-scale deployment and the operating parameters in the form of a White Space Map
of TVWS-capable systems, thereby benefitting millions of (WSM) from either the GDD-Enabling station or the
people unable to enjoy Internet due to socio-economic or Registered Location Secure Server (RLSS), the validity of
topological hurdles. Furthermore, performance analyses of which is confirmed by the Contact Verification Signal (CVS)
IEEE 802.11af prototypes are presented in [10] and [11]. In from the GDD-Enabling station. The channel availability and
addition to the above research portraying White Space the WSM are shared between the GDD-Enabling station and
availability assessments and performance analysis-based the GDD-Dependent station via the Registered Location Query
modelling, [9] presents an authentication scheme for TV WSDs Protocol (RLQP).
in order to prevent their interference with incumbent users
(primary users of the TV spectrum – terrestrial TV broadcast). The RLSS is a local database containing the permissible
frequencies and operating parameters for a group of BSSs. The
GDB-Enabling station and the RLSS collect white space
B. IEEE 802.11af Specifications: An Overview
availability information from the GDB over the Internet
The IEEE 802.11af standard [2] lays down technical (outside the scope of 802.11af). The RLSS then distributes the
specifications for intelligent shared-spectrum operations among channel availability, permissible frequencies and the operating
unlicensed White Space Devices (WSDs) and licensed parameters to the GDD-Enabling stations within its group of
(primary/incumbent) services in the TV bands. Appropriate BSSs through the established infrastructure network. Bi-
Spectrum sharing is conducted through the stringent regulation Directional over-the-air communication takes place between
of unlicensed WSD operations based on a Geo-location Data the GDD-Enabling station and the GDD-Dependent stations
Base (GDB). The IEEE 802.11af lays down certain regulations over the TV White Space bands or the ISM bands.
to ensure controlled use of these white spaces by secondary
users. The primary licensed services in the TV bands include In order to obtain the available frequencies and related
terrestrial TV services, PMSE (Programme Making and Special operational parameters, the GDD-Dependent stations launch a
Events), etc. As per the IEEE 802.11af framework, these request to the GDD-Enabling station. This inquiry is termed as
incumbent users are protected from interference within their a Channel Availability Request (CAQ). The RLSS accesses the
service area and hence unlicensed WSDs are not permitted to GDB for channel availability information for that specific
interfere with the operation of the licensed incumbent users of region and forwards it to the requesting station. The GDD-
the TV White Spaces. Dependent stations generate a CAQ when the GDD-Enabling
stations indicate a channel availability change via a CVS or
The propitious propagation characteristics of signals in the when the GDD-Dependent station has ventured out beyond the
TV White Spaces (VHF/UHF spectrum) make them highly regulatory geographic region or when the session enablement
sought after for a variety of consumer applications. Firstly, time has expired and the station wishes to remain in the GDD
since the TV bands reside in the sub-GHz spectrum, material Enabled state.
obstruction is significantly lesser than in the unlicensed ISM
(2.4GHz – IEEE 802.11b, 802.11g, 802.11n WLAN standards) A Channel Schedule Management (CSM) request is
and UNII (5GHz – IEEE 802.11a, 802.11ac WLAN standards performed by the GDD-Enabling station to the RLSS or other
with DFS and TPC) bands. Secondly, the TV bands exhibit GDD-Enabling stations, in order to acquire information about
significant path loss advantage over the ISM and UNII bands changes in channel schedules, channel usage start/stop times, et
cetra. Once the GDD-Dependent STAs receive the WSM from
the GDD-Enabling stations the WSM has to be validated. In Doordarshan, the sole broadcaster in the country’s
order to validate the WSM, the components of the 802.11af terrestrial TV spectrum is slowly phasing out analog
framework make use of the Map ID field in the Contact transmissions and hopes to completely digitize its terrestrial
Verification Signal (CVS). If the Map ID field in the CVS TV transmissions by 2017. This opens up even greater
matches the existing WSM, the GDD-Dependent station opportunities for unlicensed WSDs to operate in these bands.
assumes validation of the WSM by the GDD-Enabling station Doordarshan (comprising two All-India channels – DD News
and resets its enablement validation timer. But if the WSM ID and DD National) transmits in the VHF Band-I (54-68 MHz: 2
does not match the received CVS Map ID, then the GDD- channels of 7MHz each), VHF Band-III (174-230MHz: 8
Dependent station initiates a CAQ to the GDD-Enabling station channels of 7MHz each) and UHF Band-IV (470-582MHz: 14
to obtain the correct operating parameters in the form of a channels of 8MHz each). Out of these three operating bands
WSM. However, if the GDD-Dependent station fails to obtain which collectively constitute 1415 terrestrial transmitters all
the valid WSM before the enablement validation timer expires, over India, we primarily focus on UHF Band-IV (470-
the GDD-Dependent station will stop all transmissions. 582MHz) for the prototyping of our system.
The GDD-Dependent stations have three operational states
namely, Un-enabled, Attempting Enablement and Enabled. In III. SYSTEM ARCHITECTURE: OVERALL BLOCK DIAGRAM
the Un-enabled state, the GDD-Dependent station is not
authorized to start/resume transmissions. Instead, the GDD-
Dependent station simply scans the spectrum for a GDD-
Enablement signal from a GDD-Enabling station.
Once the GDD-Dependent station receives a GDD-
Enablement signal, it enters into the Attempting GDD
Enablement state and is only authorized to send the GDD-
Enabling response so as to complete the GDD authentication
and channel access procedure. If the GDD-Dependent station
receives a successful GDD-Enablement response from the
GDD-Enabling station, it enters into the GDD-Enabled state
and the enablement validation timer starts. The state of the
GDD-Dependent station can be changed from GDD-Enabled to
un-enabled by either the expiry of the timer or upon receipt of
an Authorization Disabled frame by the specific GDD-
Enabling station.
Between the GDD-Enabling station and the GDD-
Dependent station a Network Channel Control (NCC)
procedure takes place. The NCC requesting station (GDD-
Dependent station) requests for the use of specific frequencies
from its White Space Map by providing their spectral mask.
Upon receiving this request, the NCC-responding station
Fig. 1. Architecture of the proposed ‘Hardware-in-Loop’ system
(GDD-Enabling station) responds by providing the valid
network channels and related transmit power provided the
requested channels are available for use. A. Block Diagram Description
The proposed system consists of a UHF Scanner, a
The IEEE 802.11af standard also lays down specifications Network Processor, a UHF Translator and a Wi-Fi IEEE
regarding the PHY and MAC layer procedures for successful 802.11b/g/n source. The UHF scanner consists of a commercial
TVWS operation. The TV High Throughput (TVHT) spectrum analyzer equipped with a UHF antenna to scan the
specification in the IEEE 802.11af standard stipulates the use terrestrial TV spectrum and develop the GDB to facilitate
of 144 OFDM subcarriers for 6MHz and 8MHz channels and compliance with the IEEE 802.11af standard and ensure non-
168 OFDM subcarriers for 7MHz channels. The TVHT interference with incumbent users of the TV spectrum. The
specification supports single channel bandwidth or Basic Network Processor is the brains behind the operation, i.e. it
Channel Unit of 6,7 or 8MHz and it also provides for the use of controls the operation and operating parameters of the system
MIMO transmissions with 4x STBC and 4x MU diversity. based on data from the established GDB.
Analyzing the potential of White Spaces in the Terrestrial The Wi-Fi Radio source can either be a commercial Wi-Fi
TV spectrum in India sheds some light on the relevance and radio module compliant with IEEE WLAN 802.11b/g/n
viability of our system. The WPC, MCIT, Govt. of India standards or a software-simulated/ recorded source.
published the National Frequency Allocation Plan (NFAP)
which provides a synoptic view of the possible bands of The UHF Translator subsystem up-converts the UHF
interest for potential White-Fi broadband solutions. The NFAP- uplink signals to the ISM band (2.4GHz) and down-converts
2011-WPC allocates 470-585MHz for Fixed, Mobile and the ISM band downlink signals to the UHF band whenever the
Broadcasting services. network processor grants enablement of TVWS transmissions
based on authorization from the GDB.
 The prototype discussed in this article is a Hardware-in- B. Simulation of the UHF Translator in MATLAB
Loop processing system employing a consolidated UHF The UHF Translator subsystem is simulated in MATLAB
scanner, Network Processor, UHF Translator and the Wi-Fi based on the concepts of Digital Up-conversion and Digital
source. The UHF scanner employed in the prototype is the
Down-conversion. Numerically Controlled Oscillators (NCOs)
Keysight EXA N9010A spectrum analyzer equipped with the
set to required frequencies are employed for the corresponding
Diamond RH799 UHF Telescopic Whip Antenna used to scan
the Terrestrial UHF TV spectrum in India ranging from 470- frequency translation. The input source to the developed
582MHz comprising 14 channels with a channel bandwidth of MATLAB model is an 802.11n 2.4GHz Wi-Fi signal. The
8MHz. Based on our scans at the BMSCE, Bengaluru campus DDC module consists of a three-stage filter cascade: CIC
we have developed probabilistic models to determine the Decimator, CIC Compensator and a third-stage FIR filter.
White-Space availability, permissible frequencies and Figure 2 shows the DDC Filter cascade magnitude response.
operating parameters in order to establish the Geo-Location The DDC module down-samples the Wi-Fi signal and down-
Data Base which indicates the channel availability and allowed converts it. The DUC module consists of a three-stage filter
frequencies subject to certain constraints such as geographic cascade: FIR Interpolator, CIC Compensator and CIC
location, regulatory domains, start and end times, etc. Interpolator. Figure 3 shows the DUC Filter cascade
magnitude response. In addition to the proposed Hardware-in-
The Network Processor in the proposed Hardware-in-Loop
system is MATLAB. The MATLAB Instrument Control Loop processing of the TVWS system, the aforementioned
Toolbox is used to connect to the EXA N9010A over a MATLAB simulation enabled much-needed validation of the
standard TCP/IP interface. The MATLAB script connects to operating parameters of our system.
the EXA N9010A and completely automates the operation of
the UHF scanner, i.e. defining start and stop frequencies
(specifying the band of interest), defining mode of operation
and querying the power levels in each of the 14 channels. The
automation of the Keysight EXA N9010A over LAN is done
using SCPI commands. SCPI (Standard Commands for
Programmable Instruments) is a programming language
standard developed specifically for controlling instruments. It
defines how we communicate with test and measurement
instruments from an external computer. Using a wide range of
SCPI commands to automate the operation of the UHF scanner,
the channel power levels are queried by the MATLAB script
periodically to detect any activity on these channels. The
primary purpose of the Network Processor subsystem is to
ensure non-interference of our system with licensed users. If
the channel is being used by a licensed primary user, the
Network Processor informs the UHF Translator subsystem to
immediately stop all transmissions or provide the UHF Fig. 2. DDC Filter Cascade Magnitude Response
Translator with a different free channel to enable seamless
connectivity while ensuring compliance with IEEE 802.11af
standard.
The Wi-Fi source employed for the proposed Hardware-in-
Loop system is an 802.11b Wi-Fi signal generated in the
Agilent VSA 89600B software which is then exported to the
UHF Translator subsystem developed in Agilent SystemVue.
The UHF Translator subsystem is the primary design
constraint in the proposed prototype and hence extensive
simulation-based verification of the UHF Translator subsystem
is done both in MATLAB and Agilent SystemVue. However,
broadband up/down converting wideband mixers with
integrated VCOs can be employed to carry out the role of the
UHF Translator provided an appropriate interface is employed
between the Network Processor and the UHF Translator. The
second phase of our project deals with the development of a
comprehensive hardware platform using FPGAs and
programmable wideband RF modules, to utilize the TV White
Fig. 3. DUC Filter Cascade Magnitude Response
spaces in India using a Wi-Fi radio. However, the FPGA-based
White-Fi hardware platform is beyond the scope of this article.
C. Hardware-in-Loop Processing: In-Depth Analysis For the conversion of signal frequency from 2412 MHz to
473 MHz or vice versa, multiplication of the signal in time-
domain is employed. This can be explained by using two
simple cosine signals.
x (t) = A * cos (2π * f1 * t) and y (t) = B * cos (2π * f2 * t) (1)
We know that, cos (a) * cos (b) = ½ [cos (a + b) + cos (a - b)] (2)
Therefore, z (t) = x (t) * y (t) (3)
z(t) = (AB/2)*{cos (2π * (f1 + f2) *t) + cos (2π * (f1 - f2) *t)} (4)
Since all signals can be expressed as sum of sine and cosine
signals using Fourier series, this formula holds good for all
signals. If one frequency is 2412 MHz, to get 473 MHz, we
need to use an oscillator of frequency 1939 MHz (2412 – 1939
Fig. 4. SystemVue HIL system prototype design = 473): Low side Injection LO.

The SystemVue model of the system consists of an 802.11b The frequency of oscillator is defined as ‘vary_freq’. This
Wi-Fi signal as the Wi-Fi source. The 802.11b signal is variable is recognized and allotted a specific value after the
centered to 2412MHz and a Bandwidth of 20MHz using debugging of equation. The signal is downloaded to EXG
Chebyshev second order bandpass filter. The filtered signal is Vector Signal Generator using SignalDownloader E4438C.
amplified using the amplifier model from SystemVue The integration between SystemVue and EXG Vector Signal
Algorithm Design library. The filtered and amplified Wi-Fi Generator is done using LAN connection. Using primary IP
signal is now down-converted to the ‘free’ TV band based on address of EXG, the signal can be downloaded to the EXG
authorization from the Network processor. Using the UHF from SystemVue and it can be used as a source for further
scanning subsystem described earlier and applying power processes. When the signal is sent to the EXG, it will save the
distribution analysis to the observed recordings we were able to signal as .bin file and can be used to transmit using an N-type
build probabilistic models taking into consideration the channel connector provided UHF antenna.
power levels, spurious emissions, duration of the transmissions, The UHF Scanner subsystem continuously monitors the
time of the day, etc. These models were then extrapolated to spectrum; whenever a licensed signal appears in the band, it
develop the Geo-location Data Base. The established GDB will command to change the carrier to next available channel.
serves as the repository of permissible frequencies and This process is done by equation part of the schematic design
operating parameters for WSDs requesting access to the in SystemVue. Here SCPI commands are used to take data
available white spaces in the terrestrial TV spectrum. from the Spectrum Analyzer, debug the results from Spectrum
The Network Processing subsystem consists of the Analyzer and change the frequency of Oscillator to change the
MATLAB scripts, MATLAB and EXA N9010A instrument carrier frequency (using ‘vary_freq’), so that the transmission
interfacings and the MATLAB and SystemVue interfacings. takes place in the next available channel.
The MATLAB scripts establish a TCP/IP connection with the The IEEE 802.11b Wi-Fi signal from the Agilent VSA
UHF scanner and query the channel power levels in real-time 89600B recording is shown in Figure 5. The down-converted
to detect any unforeseen activity on the concerned channels. signal based on Network Processor commands is shown in
The channel power levels are queried by MATLAB using SCPI Figure 6.
commands over the TCP/IP interface. The channel power
levels received from the EXA N9010A are then compared with
the set threshold to determine the availability of the concerned
channels. If any unforeseen activity happens on the channel,
the MATLAB script informs the UHF Translator subsystem to
cease all transmissions or switch to the next available channel.
The MATLAB-SystemVue interface is the crucial design
constraint in order to ensure compliance with the IEEE
802.11af standard, i.e. non-interference with the licensed users
of the terrestrial TV spectrum. The UHF Translator subsystem,
upon authorization from the Network Processor (i.e. in
802.11af case, GDD-enablement to switch to GDD-enabled
state which starts the enablement validation timer) converts the
Wi-Fi signal to the specific TV band.
Fig. 5. IEEE 802.11b Wi-Fi signal
 numerous test cases. Throughout the development of the
system, the prototype has been designed to be compliant with
the IEEE 802.11af standard.

ACKNOWLEDGMENT
We express our profound gratitude to [Link] Babu,
Principal, BMSCE, Bengaluru for his constant support and
encouragement. We are greatly thankful to [Link],
BMSCE, Bengaluru and TEQIP, BMSCE for their monetary
and logistical support without which this undertaking could not
have succeeded. We are greatly indebted to Mr. Anand Jain,
Applied RealTech Systems, Bengaluru for his assistance with
respect to the scanning subsystem of the prototype. Finally, we
are thankful to the Laboratory faculty of the Wireless
Communications Lab, BMSCE, Bengaluru for their assistance
with the Test & Measurement equipment.
Fig. 6. The Down-converted signal (in the authorized free channel) viewed
in XSA of the EXA N9010A REFERENCES
[1] McKinsey & Company, Technology, Media and Telecom Practice,
IV. CONCLUSION “Offline and Falling Behind: Barriers to Internet Adoption”, October
2014.
The paper throws valuable insight on the barriers impeding [2] A. B. Flores, R. E. Guerra, E. W. Knightly, P. Ecclesine and S. Pandey,
one hundred percent Internet adoption in India. The primary "IEEE 802.11af: a standard for TV white space spectrum sharing," in
cause for a huge offline population in India is the lack of IEEE Communications Magazine, vol. 51, no. 10, pp. 92-100, October
2013.
domain-specific and adjacent infrastructure. Most of the offline
population resides in economically backward and [3] P. Kumar et al., "White space detection and spectrum characterization in
urban and rural India," World of Wireless, Mobile and Multimedia
geographically isolated rural communities. We aim to reach Networks (WoWMoM), 2013 IEEE 14th International Symposium and
these rural communities and provide them with basic Internet Workshops on a, Madrid, 2013, pp. 1-6.
services through the deployment of our system. Our system [4] G. Naik, S. Singhal, A. Kumar and A. Karandikar, "Quantitative
ensures the provision of basic Internet services to these rural assessment of TV white space in India," Communications (NCC), 2014
communities thereby bringing them into the limelight, i.e. Twentieth National Conference on, Kanpur, 2014, pp. 1-6.
bringing them into the Government of India’s Digital India [5] O. Holland et al., "A series of trials in the UK as part of the Ofcom TV
campaign and facilitating seamless access to content, services white spaces pilot," Cognitive Cellular Systems (CCS), 2014 1st
International Workshop on, Germany, 2014, pp. 1-5.
and social media. Our dream is to see a Connected India which
[6] A. Saeed, M. Ibrahim, K. A. Harras and M. Youssef, "Toward dynamic
would undoubtedly accentuate the social, economic and real-time geo-location databases for TV white spaces," in IEEE
cultural landscape in the country. Network, vol. 29, no. 5, pp. 76-82, September-October 2015.
The availability of White Spaces in the Terrestrial TV [7] P. Palka and U. Reimers, "Allocation of Dynamic TV White Spaces in a
Dynamic Broadcast system," Consumer Electronics - Berlin (ICCE-
spectrum in India is validated by our surveys at the BMSCE Berlin), 2012 IEEE International Conference on, Berlin, 2012, pp. 126-
campus in Bengaluru, Karnataka. The survey resulted in the 130.
development of probabilistic models based on Power [8] M. Deshmukh, K. P. Patil, S. Kharote, A. Jadhav and S. Maskar,
Distribution Analysis which in turn led to the establishment of "Quantitative assessment of TV white spaces for cognitive radio
a comprehensive Geo-Location Data Base. networks in India," Contemporary Computing and Informatics (IC3I),
2014 International Conference on, Mysore, 2014, pp. 619-624.
Based on data from the GDB, the Network Processor [9] J. Paavola and A. Kivinen, "Device authentication architecture for TV
(similar to the GDD-Enabling station) grants authorization to white space systems," 2014 9th International Conference on Cognitive
the UHF Translator system to start/continue transmissions in Radio Oriented Wireless Networks and Communications
the authorized channels for seamless Internet access to the (CROWNCOM), Oulu, 2014, pp. 460-465.
requesting WSDs (similar to GDD-Dependent stations). [10] H. Sawada et al., "Path Loss and Throughput Estimation Models for an
IEEE 802.11af Prototype," 2015 IEEE 81st Vehicular Technology
Focusing primarily on the UHF Translator subsystem, Conference (VTC Spring), Glasgow, 2015, pp. 1-5.
MATLAB model based simulations and exclusive SystemVue [11] M. C. Macit, H. Şenol and S. Erküçük, "Performance investigation of
RF-simulations were done to validate and finalize the design. IEEE 802.11af systems under realistic channel conditions," 2015
International Wireless Communications and Mobile Computing
The major subject matter of the article is the Hardware-in-Loop Conference (IWCMC), Dubrovnik, 2015, pp. 431-435.
processing of the White-Fi system using industry-standard test
[12] R. Almesaeed, N. F. Abdullah, A. Doufexi and A. R. Nix, "A
and measurement equipment, SystemVue, MATLAB and their Throughput Study of White-Fi Networks in Rural Environment under
associated interfaces. The outputs of the HIL processing are Realistic Conditions and Mobility," 2015 IEEE 81st Vehicular
tested and observed on the EXA N9010A XSA software for Technology Conference (VTC Spring), Glasgow, 2015, pp. 1-5.